OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity
95,310 Commits 8,304 Branches 150 Tags 1.5 GiB
master
Commit Graph

1105 Commits

Author SHA1 Message Date
Jeff Daily
239e7b541a [ROCm][CI] upgrade nightly wheels to ROCm 7.1 (#166730) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/166730
Approved by: https://github.com/jeffdaily

Co-authored-by: Jeff Daily <jeff.daily@amd.com>
 2025-10-31 17:30:47 +00:00
Xuehai Pan
69be99ee51 Remove manually synced arch versions in tools/nightly.py (#166616) 
Discussed with @atalman offline. To reduce duplicate changes and reduce the number of files to change when updating arch versions.

------

Pull Request resolved: https://github.com/pytorch/pytorch/pull/166616
Approved by: https://github.com/ezyang
 2025-10-31 15:11:28 +00:00
Aaron Gokaslan
96b61844a7 [BE]: Update nvshmem to 3.4.5 (#164046) 
Release notes can be found here: https://docs.nvidia.com/nvshmem/release-notes-install-guide/release-notes/release-3405.html main difference is the addition of a CPU assisted IBGDA fallback which should allow NVSHMEM IBGDA to work on way more systems without admin intervention and without using GDRCopy.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164046
Approved by: https://github.com/ezyang, https://github.com/kwen2501
 2025-10-29 07:32:05 +00:00
Ting Lu
544b443ea1 [CD] Upgrade to CUDA 13.0.2 for nightly binaries (#165470) 
13.0.U2 is posted, adding to nightlies
Why we want to upgrade: CUDA 13.0.U2 included a new release from cuBLAS that
1. Enabled opt-in fixed-point emulation for FP64 matmuls (D/ZGEMM) which improves performance and power-efficiency.
2. Improved performance on NVIDIA [DGX Spark](https://www.nvidia.com/en-us/products/workstations/dgx-spark/) for FP16/BF16 and FP8 GEMMs.
3. adds BF16x9 FP32 emulation support for SYRK and HERK routines.
Reference: https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html#cublas-release-13-0-update-2

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165470
Approved by: https://github.com/atalman
 2025-10-28 15:14:43 +00:00
PyTorch MergeBot
74336f8c77 Revert "[CD] Upgrade to CUDA 13.0.2 for nightly binaries (#165470)" 
This reverts commit 5e769ff867.

Reverted https://github.com/pytorch/pytorch/pull/165470 on behalf of https://github.com/atalman due to Sorry reverting for now, to restore trunk health ([comment](https://github.com/pytorch/pytorch/pull/165470#issuecomment-3454166879))
 2025-10-28 02:21:48 +00:00
Ting Lu
5e769ff867 [CD] Upgrade to CUDA 13.0.2 for nightly binaries (#165470) 
13.0.U2 is posted, adding to nightlies
Why we want to upgrade: CUDA 13.0.U2 included a new release from cuBLAS that
1. Enabled opt-in fixed-point emulation for FP64 matmuls (D/ZGEMM) which improves performance and power-efficiency.
2. Improved performance on NVIDIA [DGX Spark](https://www.nvidia.com/en-us/products/workstations/dgx-spark/) for FP16/BF16 and FP8 GEMMs.
3. adds BF16x9 FP32 emulation support for SYRK and HERK routines.
Reference: https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html#cublas-release-13-0-update-2

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165470
Approved by: https://github.com/atalman
 2025-10-28 00:21:47 +00:00
Huy Do
9095a9dfae [CD] Apply the fix from #162455 to aarch64+cu129 build (#165794) 
When trying to bring cu129 back in https://github.com/pytorch/pytorch/pull/163029, I mainly looked at https://github.com/pytorch/pytorch/pull/163029 and missed another tweak coming from https://github.com/pytorch/pytorch/pull/162455

I discover this issue when testing aarch64+cu129 builds in https://github.com/pytorch/test-infra/actions/runs/18603342105/job/53046883322?pr=7373.  Surprisingly, there is no test running for aarch64 CUDA build from what I see in 79a37055e7.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165794
Approved by: https://github.com/malfet
 2025-10-18 04:16:24 +00:00
Yuanyuan Chen
e925dfcc6b Enable all SIM rules except disabled ones (#164645) 
`SIM` rules are useful for simplifying boolean expressions and enhances code readability.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164645
Approved by: https://github.com/ezyang, https://github.com/mlazos
 2025-10-17 07:27:11 +00:00
Huy Do
6dedd34c31 [CD] Skip 12.9 build on Windows (#165665) 
Per title

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165665
Approved by: https://github.com/Camyll, https://github.com/malfet
 2025-10-16 19:11:27 +00:00
Catherine Lee
64699b8042 [trymerge] Do not check for rules when reverting (#165342) 
Why do we need to check for merge rules when reverting?
Pull Request resolved: https://github.com/pytorch/pytorch/pull/165342
Approved by: https://github.com/malfet
 2025-10-13 19:07:00 +00:00
Catherine Lee
684df93975 [CI] Default keep-going true for tags of form ciflow/something/commitsha (#165180) 
Tags of the form `ciflow/something/commitsha` are usually created by running the workflow from HUD

Pull Request resolved: https://github.com/pytorch/pytorch/pull/165180
Approved by: https://github.com/huydhn
 2025-10-13 16:12:37 +00:00
Huy Do
4400c5d31e Continue to build nightly CUDA 12.9 for internal (#163029) 
Revert part of https://github.com/pytorch/pytorch/pull/161916 to continue building CUDA 12.9 nightly

Pull Request resolved: https://github.com/pytorch/pytorch/pull/163029
Approved by: https://github.com/malfet
 2025-10-11 08:26:47 +00:00
Nikita Shulga
90c0825e2d [GHF] Allow reverts from pytorch-auto-revert app (#164911) 
This is a bit weird, but author_login is not a unique field, but author_url is.

Explicitly allow https://github.com/apps/pytorch-auto-revert to issue revert commands

Update mocks by running
```
sed -i -e s/8e262b0495bd934d39dda198d4c09144311c5ddd6cca6a227194bd48dbfe7201/47860a8f57a214a426d1150c29893cbc2aa49507f12b731483b1a1254bca3428/ gql_mocks.json
```

Test plan: Run
```python
from trymerge import GitHubPR
pr=GitHubPR("pytorch", "pytorch", 164660)
print(pr.get_last_comment().author_url, pr.get_comment_by_id(3375785595).author_url)
```
that should produce
```
https://github.com/pytorch-auto-revert https://github.com/apps/pytorch-auto-revert
```
Plus added a regression test that checks two particular comments for revert validity

`pytorch-auto-revert` user is my alter ego :)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164911
Approved by: https://github.com/jeanschmidt
 2025-10-08 15:15:45 +00:00
Wei Wang
773c6762b8 [CD][CUDA13][NCCL] Fix nccl version typo for cu13 (#164383) 
https://pypi.org/project/nvidia-nccl-cu13/#history does not have 2.27.5 but 2.27.7+.
Companion PR: https://github.com/pytorch/pytorch/pull/164352

Fixes a potential binary breakage due to non-existence of referenced NCCL cu13 version.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164383
Approved by: https://github.com/tinglvv, https://github.com/Skylion007, https://github.com/atalman
 2025-10-01 21:32:25 +00:00
Ivan Zaitsev
1288c6d8bb Enable keep-going for trunk tags (#164307) 
Tags like `trunk/{sha}` are used to re-run signals by [autorevert project](https://github.com/pytorch/test-infra/blob/main/aws/lambda/pytorch-auto-revert/README.md).

We need to have `keep-going` enabled for those reruns, so that they surface all test failures, not just the first one.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164307
Approved by: https://github.com/clee2000
 2025-10-01 17:21:43 +00:00
albanD
2610746375 Revert nccl upgrade back to 2.27.5 (#164352) 
Revert https://github.com/pytorch/pytorch/pull/162351 as it breaks H100
Pull Request resolved: https://github.com/pytorch/pytorch/pull/164352
Approved by: https://github.com/atalman, https://github.com/malfet
 2025-10-01 15:27:40 +00:00
Nikita Shulga
5a93f00c79 [CI] Delete binary smoke workflows (#164260) 
Those were very useful in the past, because:
- CI builder jobs did not generates wheels, but rather run `python setup.py develop` and shared docker layers, which is no longer the case, all CI jobs produce wheels
- CD jobs were targeting pre-CXX11 ABI, but this is no longer the case after manylinux2_28 migration

Existing, but acceptable gaps:
 - Windows libtorch debug builds sometimes might fail, but IMO it's ok not to be able to produce those for a few days, as number of libtorch users are somewhat small
 - All CD jobs are based on AlmaLinux, while CI are based on Ubuntu, but this could be adjusted if needed, besides AlmaLinux-9 and Ubuntu-22.04 are pretty close in terms of glibc and gcc versions
 - CD jobs build for all GPU architectures, while CI only for the one being tested, but there are now periodic H100 and B200 jobs, and not a lot of development happens for Voltas or Pascals

Besides there are better tools to alert about the nightly failures

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164260
Approved by: https://github.com/seemethere, https://github.com/atalman
 2025-09-30 20:00:07 +00:00
Aaron Gokaslan
5504a06e01 [BE]: Update NCCL to 2.28.3 (#162351) 
@eqy New NCCL has some a bunch of bugfixes for features including reducing the number SMs needed by NVLINK collectives as well as some very useful new APIs for SymmetricMemory.  Also allows FP8 support for non-reductive operations on pre-sm90 devices.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162351
Approved by: https://github.com/ezyang, https://github.com/malfet, https://github.com/atalman
 2025-09-28 01:38:59 +00:00
Jeff Daily
f1260c9b9a [ROCm][CI/CD] upgrade nightly wheels to ROCm 7.0 (#163937) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/163937
Approved by: https://github.com/jeffdaily

Co-authored-by: Jeff Daily <jeff.daily@amd.com>
 2025-09-26 21:42:09 +00:00
Nikita Shulga
00f96dd84d [CI] Run CUDA-13 binary builds on trunk (#163787) 
There are numerous other workflows that could be used to catch CUDA-12
build regression (our CI builds are almost identical to CD ones), but not many CUDA-13 builds around, so https://github.com/pytorch/pytorch/issues/163342 are really hard to detect in CI
Pull Request resolved: https://github.com/pytorch/pytorch/pull/163787
Approved by: https://github.com/atalman, https://github.com/huydhn
 2025-09-25 00:58:17 +00:00
Nikita Shulga
52dd7a898c Move ROCM trunk wheel builds to 3.10 (#163339) 
This code is a delicious spaghetti: Sometimes python version is defined in jinja template (see https://github.com/pytorch/pytorch/pull/162297 ) sometimes in shell script (see https://github.com/pytorch/pytorch/pull/162877 ), but this time around it's in a python file (and there is another one called `generate_binary_build_matrix.py` that defines `FULL_PYTHON_VERSIONS`)

Pull Request resolved: https://github.com/pytorch/pytorch/pull/163339
Approved by: https://github.com/clee2000
 2025-09-19 18:52:00 +00:00
Huy Do
66133b1ab7 Build vLLM aarch64 nightly wheels (#162664) 
PyTorch has published its aarch64 nightly wheels for all CUDA version after https://github.com/pytorch/pytorch/pull/162364
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162664
Approved by: https://github.com/atalman
 2025-09-13 03:43:55 +00:00
Svetlana Karslioglu
e15686b40d Remove actionable label from docathon label sync script (#155713) 
Make sure we don't propagate actionable label in docathon sync label script.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/155713
Approved by: https://github.com/clee2000
 2025-09-12 15:36:50 +00:00
Ting Lu
bb1d53bc47 [CD] CUDA 13 specific followup changes (#162455) 
Follow up for CUDA 13 bring up https://github.com/pytorch/pytorch/issues/159779
sm50-70 should not be added to sbsa build arch list, as previous archs had no support for arm.
remove platform_machine from PYTORCH_EXTRA_INSTALL_REQUIREMENTS

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162455
Approved by: https://github.com/atalman
 2025-09-11 00:03:47 +00:00
Huy Do
e64965300a Repackage vLLM nightlies (#162371) 
I suspected that I would need to repack vLLM wheels from https://github.com/pytorch/pytorch/pull/162000 because I renamed the wheel, and it turns out to be true.  The error is as follows:

```
$ uv pip install --pre xformers --index-url https://download.pytorch.org/whl/nightly/cu129
Using Python 3.12.11+meta environment at: venv/py3.12
Resolved 28 packages in 759ms
error: Failed to install: xformers-0.0.33.dev20250901+cu129-cp39-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (xformers==0.0.33.dev20250901+cu129)
  Caused by: Wheel version does not match filename: 0.0.33+5d4b92a5.d20250907 != 0.0.33.dev20250901+cu129
```

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162371
Approved by: https://github.com/atalman
 2025-09-10 04:02:34 +00:00
Ke Wen
8922bbcaab Use same NVSHMEM version across CUDA builds (#162206) 
#161321 bumped NVSHMEM version to 3.3.24 for CUDA 13, leaving CUDA 12 with 3.3.20.
This PR bumps the NVSHMEM version to 3.3.24 for CUDA 12 as well.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162206
Approved by: https://github.com/tinglvv, https://github.com/Skylion007
 2025-09-09 20:59:50 +00:00
PyTorch MergeBot
5ccf3ca3ec Revert "Use same NVSHMEM version across CUDA builds (#162206)" 
This reverts commit 0d9c95cd7e.

Reverted https://github.com/pytorch/pytorch/pull/162206 on behalf of https://github.com/malfet due to Broke lint, see 4dd73e659a/1 ([comment](https://github.com/pytorch/pytorch/pull/162206#issuecomment-3271040521))
 2025-09-09 14:40:45 +00:00
Ke Wen
0d9c95cd7e Use same NVSHMEM version across CUDA builds (#162206) 
#161321 bumped NVSHMEM version to 3.3.24 for CUDA 13, leaving CUDA 12 with 3.3.20.
This PR bumps the NVSHMEM version to 3.3.24 for CUDA 12 as well.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162206
Approved by: https://github.com/tinglvv, https://github.com/Skylion007
 2025-09-09 08:52:27 +00:00
Ting Lu
9c991b63ff [CD] [aarch64] Add CUDA 12.6 and 12.8 to build matrix, remove 12.9 build (#162364) 
https://github.com/pytorch/pytorch/issues/159779

Add the full CUDA support matrix to sbsa build (12.6, 12.8)
Same arch support as x86 build
Remove 12.9 sbsa build

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162364
Approved by: https://github.com/atalman
 2025-09-08 20:00:25 +00:00
David Berard
3f5993316e [upstream triton] update triton pin to triton 3.5 (#162278) 
Update PyTorch to the latest Triton release candidate branch (release/3.5.x in triton-lang/triton)

Notably:
* this does *not* include the version number bump from 3.4 -> 3.5 (we'll do that in a follow-up PR)
* sam_fast is still failing, so we've disabled it temporarily https://github.com/pytorch/pytorch/issues/162282 and we are committed to fixing it, ideally before the branch cut but possibly as a cherry-pick into the release branch.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162278
Approved by: https://github.com/atalman
ghstack dependencies: #162244, #162309
 2025-09-08 14:29:24 +00:00
Eddie Yan
145a3a7bda [CUDA 13][cuDNN] Bump CUDA 13 to cuDNN 9.13.0 (#162268) 
Fixes some `d_qk` != `d_v` cases on Hopper that are broken by cuDNN 9.11-9.12

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162268
Approved by: https://github.com/drisspg, https://github.com/Skylion007
 2025-09-06 01:59:03 +00:00
atalman
bffc7dd1f3 [CD] Add cuda 13.0 libtorch builds, remove CUDA 12.9 builds (#161916) 
Related to https://github.com/pytorch/pytorch/issues/159779

Adding CUDA 13.0 libtorch builds, followup after https://github.com/pytorch/pytorch/pull/160956
Removing CUDA 12.9 builds, See https://github.com/pytorch/pytorch/issues/159980

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161916
Approved by: https://github.com/jeanschmidt, https://github.com/Skylion007

Co-authored-by: Ting Lu <tingl@nvidia.com>
 2025-09-05 07:47:54 +00:00
PyTorch MergeBot
6b8b3ac440 Revert "[ROCm] Use MI325 (gfx942) runners for binary smoke testing (#162044)" 
This reverts commit cd529b686d.

Reverted https://github.com/pytorch/pytorch/pull/162044 on behalf of https://github.com/jeffdaily due to mi200 backlog is purged, and mi300 runners are failing in GHA download ([comment](https://github.com/pytorch/pytorch/pull/162044#issuecomment-3254427869))
 2025-09-04 16:06:30 +00:00
Jithun Nair
cd529b686d [ROCm] Use MI325 (gfx942) runners for binary smoke testing (#162044) 
### Motivation

* MI250 Cirrascale runners are currently having network timeout leading to huge queueing of binary smoke test jobs:
<img width="483" height="133" alt="image" src="https://github.com/user-attachments/assets/17293002-78ad-4fc9-954f-ddd518bf0a43" />

* MI210 Hollywood runners (with runner names such as `pytorch-rocm-hw-*`) are not suitable for these jobs, because they seem to take much longer to download artifacts: https://github.com/pytorch/pytorch/pull/153287#issuecomment-2918420345 (this is why these jobs were specifically targeting Cirrascale runners). However, it doesn't seem like Cirrascale runners are necessarily doing much better either e.g. [this recent build](https://github.com/pytorch/pytorch/actions/runs/17332256791/job/49231006755).
* Moving to MI325 runners should address the stability part at least, while also reducing load on limited MI2xx runner capacity.
* However, I'm not sure if the MI325 runners will do any better on the artifact download part (this may need to be investigated more) cc @amdfaa

* Also removing `ciflow/binaries` and `ciflow/binaries_wheel` label/tag triggers for `generated-linux-binary-manywheel-rocm-main.yml` because we already trigger ROCm binary build/test jobs via these labels/tags in `generated-linux-binary-manywheel-nightly.yml`. And for developers who want to trigger ROCm binary build/test jobs on their PRs, they can use the `ciflow/rocm-mi300` label/tag as per this PR.

### TODOs (cc @amdfaa):
* Check that the workflow runs successfully on the MI325 runners in this PR. Note how long the test jobs take esp. the "Download Build Artifacts" step
* Once this PR is merged, clear the queue of jobs targeting `linux.rocm.gpu.mi250`

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162044
Approved by: https://github.com/jeffdaily

Co-authored-by: Jeff Daily <jeff.daily@amd.com>
 2025-09-03 18:34:07 +00:00
Aleksei Nikiforov
71992dd805 S390x: build nightly binaries for new pythons (#161920) 
Enable python 3.13t, 3.14 and 3.14t on s390x for nightly binaries

Fixes #161515

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161920
Approved by: https://github.com/malfet
 2025-09-03 17:38:38 +00:00
Ting Lu
fefee08164 [CD] Add CUDA 13.0 Windows build (#161663) 
Test CUDA 13.0 windows build

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161663
Approved by: https://github.com/malfet, https://github.com/atalman
 2025-09-01 15:27:17 +00:00
Zain Rizvi
c8fa907e74 Check commit order (#161560) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161560
Approved by: https://github.com/malfet
ghstack dependencies: #161558, #161637
 2025-08-29 16:22:58 +00:00
Zain Rizvi
6b051d7de3 [BE] Refactor trymerge for readability (#161637) 
Two changes:
- Extract getting the last_commit's sha into it's own function
- Rename merge_changes to merge_changes_locally to better explain it's functionality
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161637
Approved by: https://github.com/seemethere, https://github.com/malfet
ghstack dependencies: #161558
 2025-08-27 22:44:00 +00:00
Zain Rizvi
624bc36163 Ensure the comment id is always passed in to trymerge (#161558) 
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161558
Approved by: https://github.com/seemethere, https://github.com/malfet
 2025-08-27 19:53:28 +00:00
Wang, Chuanqi
06c7516994 [BE] Upgrade XPU support package to 2025.2 (#158733) 
Including below changes,

- Add XPU support package 2025.2 build and test in CI for both Linux and Windows
- Keep XPU support package 2025.1 build in CI to ensure no break issue until PyTorch 2.9 release
- Upgrade XPU support package from 2025.1 to 2025.2 in CD for both Linux and Windows
- Rename Linux CI job name & image name to n & n-1
- Update XPU runtime pypi packages dependencies of CD wheels
- Remove deprecated support package version docker image build

Pull Request resolved: https://github.com/pytorch/pytorch/pull/158733
Approved by: https://github.com/EikanWang, https://github.com/atalman
 2025-08-27 19:33:38 +00:00
Ting Lu
9632f4ea9f [CD] [aarch64] Add CUDA 13.0 sbsa nightly build (#161257) 
https://github.com/pytorch/pytorch/issues/159779

CUDA SBSA build for CUDA 13.0
1. Supported archs: sm_80 to sm_120. Including support for Thor (sm_110), SPARK (sm_121), GB300 (sm_103).
"This release adds support of SM110 GPUs for arm64-sbsa on Linux." from 13.0 release notes https://docs.nvidia.com/cuda/cuda-toolkit-release-notes/index.html
2. Use -compress-mode=size for binary size reduction, 13.0 wheel is 2.18 GB, when compared with 12.9 3.28 GB, that is 1.1 GB of savings and ~33.5% smaller.
3. Refactored the libs_to_copy list with common libs, and version_specific_libs.

TODO: add the other CUDA archs in the existing support matrix of x86 to SBSA build as well

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161257
Approved by: https://github.com/nWEIdia, https://github.com/atalman
 2025-08-27 14:38:07 +00:00
Ting Lu
ae8d319fd4 Update NVSHMEM to 3.3.24 and fix download link (#161321) 
https://github.com/pytorch/pytorch/issues/159779

Update NVSHMEM 3.3.24 for [PyTorch CUDA13 Binary Cannot Be Built with SM_75 with NVSHMEM](https://github.com/pytorch/pytorch/issues/160980)
Enabled back sm_75 for NVSHMEM
Fixed the NVSHMEM download link for the issue with 3.3.20 download in issue - [[CD] nvshem-3.3.9 wheels for aarch64 is not manylinux2_28 compliant](https://github.com/pytorch/pytorch/issues/160425)

Todo: Should also enable back build ARM with NVSHMEM since it is compatible with manylinux2_28

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161321
Approved by: https://github.com/Skylion007, https://github.com/atalman
 2025-08-26 13:26:18 +00:00
atalman
1a566c4909 Remove Python 3.9 nightly builds (#161427) 
Please see https://github.com/pytorch/pytorch/issues/161167

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161427
Approved by: https://github.com/huydhn
 2025-08-25 22:05:40 +00:00
Wang, Chuanqi
a43480d19c [CD] Enable triton xpu Windows build for Python 3.14 (#161255) 
Follow #159869
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161255
Approved by: https://github.com/atalman
 2025-08-22 18:39:31 +00:00
Ting Lu
49ff884b1e Add CUDA 13.0 x86 builds (#160956) 
https://github.com/pytorch/pytorch/issues/159779

CUDA 13.0.0
NVSHMEM 3.3.20
CUDNN 9.12.0.46

Adding x86 linux builds for CUDA 13.
Adding libtorch docker.
Package naming changed for CUDA 13 (removed postfix -cu13 for some packages).

Preparation checklist:
1. Update index https://download.pytorch.org/whl/nightly/cu130 with pypi packages
2. Update packaging name based on https://pypi.org/project/cuda-toolkit/ metadata

Pull Request resolved: https://github.com/pytorch/pytorch/pull/160956
Approved by: https://github.com/atalman

Co-authored-by: atalman <atalman@fb.com>
 2025-08-22 11:31:09 +00:00
Ting Lu
a68f63e331 Add Windows CUDA 13 build and magma script (#161073) 
Add magma build 13.0 for Windows
Add cuda_install.bat 13.0 for Windows build
https://github.com/pytorch/pytorch/issues/159779

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161073
Approved by: https://github.com/atalman

Co-authored-by: Andrey Talman <atalman@fb.com>
 2025-08-22 11:24:25 +00:00
Nikita Shulga
e1a64b75ff [CD] Delete full builds (#161075) 
As they are no longer needed for Colab, see https://github.com/googlecolab/colabtools/issues/5508#issuecomment-3200871941 and
[<img width="896" height="128" alt="image" src="https://github.com/user-attachments/assets/a287393c-bde7-4e10-99bf-2e0d66346efe" />
](https://colab.research.google.com/drive/1YJ5Y0xsApXSewM1cQwWQ_AS3A77vytgq)

Fixes https://github.com/pytorch/pytorch/issues/160972
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161075
Approved by: https://github.com/atalman
 2025-08-20 19:40:15 +00:00
atalman
62db8ec391 windows python 3.14 nightly builds (#159869) 
Related to https://github.com/pytorch/pytorch/issues/156856

Pull Request resolved: https://github.com/pytorch/pytorch/pull/159869
Approved by: https://github.com/malfet, https://github.com/williamwen42
 2025-08-19 18:36:16 +00:00
Nikita Shulga
7bd4cfaef4 [BE] Update nvshem dependency to 3.3.20 (#160458) 
Which is manylinux2_28 compatible, even on aarch64 platform

archive contents and URL pattern changed quite drastically between 3.3.9 and 3.3.20, but hopefully it still works.
Package `libnvshmem_host.so.3` into gigantic aarch64+CUDA wheel
Should fix https://github.com/pytorch/pytorch/issues/160425
Pull Request resolved: https://github.com/pytorch/pytorch/pull/160458
Approved by: https://github.com/Skylion007, https://github.com/kwen2501, https://github.com/nWEIdia, https://github.com/atalman, https://github.com/tinglvv
 2025-08-16 02:00:57 +00:00
PyTorch MergeBot
c015e53d37 Revert "[BE] Update nvshem dependency to 3.3.20 (#160458)" 
This reverts commit e0488d9f00.

Reverted https://github.com/pytorch/pytorch/pull/160458 on behalf of https://github.com/wdvr due to need to rerun workflow generation (failing workflow-checks) ([comment](https://github.com/pytorch/pytorch/pull/160458#issuecomment-3193133706))
 2025-08-16 01:47:42 +00:00